Natural history of untreated head and neck cancer L.P. Kowalski *, A.L. Carvalho From Head and Neck Surgery and Otorhinolaryngology Department, Centro de Tratamento e Pesquisa, Hospital do Ca Ãncer, A.C. Camargo, Rua Antonio Prudente, 211 Sa Äo Paulo, Brazil Received 20 September 1999; received in revised form 21 February 2000; accepted 22 February 2000 Abstract This paper describes the characteristics and natural history of the largest reported group of patients with untreated head and neck cancer. From 1953 to 1990, 808 untreated head and neck cancer patients were followed-up until their death. All patients were given supportive care, but speci®c oncological treatment was not pursued secondary to advanced tumour stage, poor performance status and/or patients' refusal of any treatment. The overall survival ranged from 1 day to 53.8 months (median 3.82 months). Perfor- mance status was the most signi®cant predictor of survival (P<0.001). A subgroup of 357 patients with good performance status and aged less than 70 years old had a 12.9% 1-year survival. Approximately 50% of untreated head and neck cancer patients will die within 4 months of their diagnosis. However, the remaining patients can survive up to 4 or more years, depending on their tumour location, extent, performance status and level of supportive care. # 2000 Published by Elsevier Science Ltd. All rights reserved. Keywords: Head and neck cancer; Survival; Natural history; Untreated 1. Introduction The natural history of untreated head and neck squa- mous cell carcinoma is poorly understood. This infor- mation is essential to provide the necessary background to evaluate and recommend various treatment options including supportive care, palliation or attempting curative treatment. It has been known for a long time that select patients with head and neck cancer (HNC) can survive longer than 5 years without treatment. Sur- vival results secondary to various treatment regimes must therefore be compared with the results found in non-treated head and neck cancer patients. Possible negative changes in quality of life and mini- mal prolongation of survival induced by oncological treatment, suggests that not all patients should be trea- ted. Perhaps some should receive only supportive care [1,2]. Burns and colleagues [1] discussed the lack of evi- dence for prolongation of life or even palliation with certain treatment regimes. The crucial question remains ``to treat or not to treat''. ``Is treatment cost-eective?'' ``Does treatment add toxicity and suering without any signi®cant bene®t?'' Burns and asscoiates [1] concluded that palliative treatment (radiotherapy or chemother- apy) in unresectable stage IV HNC was of limited value. No patient was alive 2 years following treatment and the mean survival time was only 8.4 months. However, to compare the results of any form of experimental, palliative or protocol studies, comparisons with the expected results of supportive care alone is essential in end-stage disease. Other authors have described prolonged survival in a few cases of untreated HNC [3±6]. However, the pri- mary purpose of this study was to analyse the prog- nostic signi®cance of demographic factors, tumour characteristics, performance status, clinical stage (CS) and supportive care, on the natural history of a large group of 808 untreated patients followed until their death. These data can be used to improve our under- standing of the course of head and neck cancer and serve as a comparison group for all treatment options. 2. Patients and methods For this study, the charts of 808 untreated head and neck cancer patients seen from 1953 to 1990, were ret- rospectively reviewed. All patients were seen at the 0959-8049/00/$ - see front matter # 2000 Published by Elsevier Science Ltd. All rights reserved. PII: S0959-8049(00)00054-X European Journal of Cancer 36 (2000) 1032±1037 www.ejconline.com * Corresponding author. Tel.: +55-11-3272-5125; fax: +55-11- 3272-5098. E-mail address: 1p_kowalski@uol.com.br (L.P. Kowalski).